Drug addiction, and cocaine addiction in particular, is considered a chronic relapsing disorder in which subjects episodically administer the drug and ultimately transition from nondependent drug use to the compulsive drug use of addiction. Knowledge of the neurochemical/neurocircuitry changes that provide the motivational basis for vulnerability for increased drug intake with extended access are beginning to provide insights into the neurobiological changes that may lead to vulnerability to escalation in drug intake and relapse. During the previous funding period an animal model of extended access to cocaine self-administration that resembles the compulsivity of addiction has been established and validated. Animals with extended access increase their drug intake (escalation) over time and show increased motivation to obtain the drug (increased progressive- ratio [PR] responding) and show reward deficits during abstinence from the drug (elevation in brain reward thresholds). Work during the previous funding period also has shown that extended access to cocaine decreases basal release in the mesolimbic dopamine system, and inhibitory G-protein function, and increased sensitivity to dopamine antagonists and partial agonists. Chronic cocaine, cocaine withdrawal, and stress- induced reinstatement of cocaine self-administration also were associated with increases in CRF, norepinephrine, and dynorphin function in the extended amygdala. The overall hypothesis under test is that the neuroplastic changes in the extended amygdala lead to a cascade of neurobiological changes that initially involve excessive dopamine release, subsequent loss of dopamine function, and subsequent activation of CRF and dynorphin brain stress systems. A subhypothesis is that the binge-induced loss of dopamine function observed with psychostimulant drugs interacts with the brain stress systems in the basal forebrain to further promote and sustain the increased cocaine intake with extended access. To test this hypothesis, the present proposal will validate a stress-induced increase in cocaine self-administration paradigm and explore the role of CRF and dynorphin in stress-induced escalation (Specific Aims 1 and 2), explore the role of decreases in dopamine function measured by Go in the dopamine system in drug and stress-induced escalation (Specific Aim 3) and explore the interaction of Go in the ventral striatum and CRF and dynorphin function in the extended amygdala in stress-induced escalation (Specific Aim 4). These studies will not only provide new data in the role of stress in the compulsivity that is associated with cocaine dependence, but may also provide key markers for the development of dependence and ultimately key targets for understanding vulnerability and developing novel treatments.